Food and diet

Probiotic pickled turnip touted as ‘flu wonder cure’

“Wonder cure for killer flu," is the headline on the front page of the Daily Express. What the headline fails to tell you is that the ‘cure’ – bacteria found in suguki, a type of pickled turnip – has only been tested in mice.

The headlines were based on a simple experiment whereby mice were given a bacterial extract called Lactobacillus brevis (KB290) found in suguki.

The mice received the extract for 14 days before they were infected with the flu virus. Those receiving the extract showed less severe symptoms of flu during the subsequent seven days. Specifically, those given the bacterial extract had less weight loss and less severe declines in general health compared to those given an inactive placebo.

However mice given the Japanese pickle extract still experienced declines in body weight of approximately 20% by day seven and significant decreases in general health. So the mice certainly weren’t cured of flu completely, they were just slightly less ill.

Also mice are not humans. If the extract was developed successfully for use in humans – a big if – we don’t know whether people who took this extract would be any less likely to get flu, or have less severe symptoms of flu. Also importantly, we don’t know whether there would be any adverse health effects from regular use of this extract.

So any talk of a "wonder cure" based on this evidence is frankly misleading.

Where did the story come from?

The study was carried out by researchers from institutes in Japan and New Zealand. No funding source was stated. Two of the seven researchers were employed by Kagome Company – a manufacturer of, among other things, vegetable drinks and probiotic products. Though the study stated that there were no conflicts of interest. 

The study was published in the peer-reviewed science journal Applied Microbiology.

The quality of some of the UK media’s reporting of this study was poor. The Daily Express’s headline (in the print version of the newspaper, not its website) that a “Wonder cure for killer flu” has been found is simply not true. And its claim that “New drink will save thousands of lives” is based on speculation rather than evidence.

The Mail Online’s coverage is slightly more restrained but it still overstates the study’s findings.

There are many more hurdles and further experiments needed before this early-stage research could potentially translate into a medicine or “cure” people can take. Many promising findings first shown in mice fail to work in humans and so do not get developed into medicines. So the media’s leap from mouse study to human treatment is a large and potentially misguided one.

What kind of research was this?

This was a mouse study investigating whether a bacteria isolated from suguki – a traditional Japanese pickled turnip – could protect against flu in mice.

Rodent studies are useful to investigate potential new treatments for disease or infection. However, potential treatments showing promise in mice don’t always work when subsequently tested in humans. So one should not assume that if effectiveness is shown in mice, it is a given that a new treatment is on its way.

Influenza, or flu for short, is an infection caused by the influenza virus that produces fever, chills, a sore throat, weakness and a runny nose, among other symptoms. The symptoms can become severe and potentially life threatening in older people, infants or others with weaker immune systems who are less able to fight off the virus.

Flu vaccination is one effective way of minimising the chance of getting flu but it is also useful, the researchers say, to look for things that might boost the immune system to promote resistance against flu in daily life. To this end the current research investigated whether bacteria isolated from suguki had any protective effects against flu in mice. Suguki has previously been reported to have positive effects on the immune system.

What did the research involve?

Bacteria called Lactobacillus brevis (KB290) extracted from suguki were given orally to mice during a 14-day period to guard against flu. After 14 days of receiving the prophylactic “treatment” the mice were infected with flu. The specific flu virus was the H1N1, variants of which led to the 1918 “Spanish flu” pandemic that killed more people than the First World War and was one of the deadliest killers in recent human history. Variants of H1N1 were also behind the more recent 2009 flu pandemic known as “swine flu”.

Mice were randomly assigned into three groups of 20:

  • Normal: administered potato starch for 14 days (an inactive treatment), and not subsequently infected with flu.
  • Control: administered potato starch for 14 days before being infected with flu.
  • KB209: administered KB209 for 14 days before being infected with flu.

The mice were infected with two large doses of flu; the authors reported that each was equivalent to half the flu dose that would kill them.

After infection the mice were monitored daily over seven days to assess any body weight changes – loss of body weight is a key characteristic of flu infection in mice. Changes in other aspects of their physical condition caused by the flu infection were also assessed and expressed as a general health score. Further tests sought to investigate the biological mechanism through which KB290 might be protecting the mice.

The analysis compared the weight change and general health scores between the three groups of mice tested.

What were the basic results?

The main results were as follows:

  • Mice in the normal group who were not infected with flu did not lose any weight or suffer ill health as measured with the general health score.
  • Mice given the inactive treatment (control group) and KB290 group both lost more than 10% of their body weight after four days of flu infection which increased to about 20% by day seven.
  • However, those given the prophylactic KB290 lost significantly less body weight than the control group. Both KB290 and control groups were similar up to day three but from days four to seven the KB290 lost significantly less weight.
  • Mirroring this trend, mice in both the KB290 and control groups suffered similar reductions in general health score up to day three but from days four to seven the KB290 group's score declined at a significantly slower rate than the controls.
  • Biological analysis showed that there were significant differences in immune response to the flu virus between mice given KB290 and the control group, suggesting a plausible biological mechanism underpinning the differences in symptoms observed.

How did the researchers interpret the results?

In light of the findings that KB290 slowed the rate of body weight loss and slowed the decline in general health scores the study authors concluded the results, “suggested that oral administration of KB290 significantly alleviated the clinical symptoms induced by IFV [flu virus] infection”.


This simple experiment showed that giving a bacterial extract found in Japanese pickled turnip to mice, as a preventative treatment for flu, reduced some of the symptomatic effects of flu upon subsequent infection. Specifically, it lessened body weight loss and reduced declines in general health during a seven-day flu period.

While the results are encouraging, it is too soon to roll out the red carpet and welcome a “new wonder cure” as both the headlines in the Daily Express and Mail Online suggest.

For instance, mice given the Japanese pickle extract still experienced declines in body weight of approximately 20% by day seven and significant decreases in general health, so they certainly weren’t cured of flu completely. However, the symptoms were significantly milder than mice that didn’t receive the extract.

Any new treatment that could potentially help those at higher health risk from severe flu infection, such as the very young, very old, or others with a weakened immune system, would be an important breakthrough. Currently available antiviral medications can only shorten the duration of a flu infection by around a day.

However, it is far too early to say that this bacterial extract is the breakthrough the world is waiting for.

This was a study in mice, so many further experiments confirming the findings are needed. Following these, and assuming they are successful, further experiments will be needed on humans and again if effective clinical trials could start. At each of these stages in treatment development there could be failure leading to no treatment. For example the effective dose of extract would have to be established in humans, as well as how often it would have to be given. Importantly, its safety needs to be evaluated against its possible benefits (which may only include reduced symptoms rather than resolving all symptoms or preventing flu infection in the first place) to make sure the benefits outweighed the risks. So, the promise in mice may not translate into a treatment in humans.

A further limitation to consider is that the mice were infected with two large doses of flu; the authors reported that each was equivalent to half the flu dose that would kill them. Humans are not infected in such an experimental way; they contract the flu virus and develop symptoms, but it is not usually known what "dose" of virus particles they have contracted.

The severity of infection, and whether it leads to complications, is likely to be determined by the underlying health and characteristics of the person – not just how many virus particles they took in.

Therefore it is not clear whether the flu dose used in the mice is unrealistically high for humans and whether the extract would be equally effective at other flu doses.

A positive side of the potential treatment is that it might be taken orally as was the case in the mice, as opposed to many animal experiments where treatments are injected. The bacteria are already found in some probiotic drinks so, if proved to be effective in humans, this would suggest it could be a relatively simple and person-friendly option.

Due to the immense commercial potential of a "flu drink" – even one with limited effects – it's highly likely that further experimentation will be forthcoming. There are, unconfirmed, media reports of ongoing human trials.

NHS Attribution